CONTROLLLING WOODY VEGETATION IN WETLAND PRARIES
1994-1999
Prepared by
Deborah L. Clark
and
Mark V. Wilson
Oregon State University
Submitted to
Coast Range Resource Area Bureau of Land Management Eugene, Oregon
Project No. HE-P99-0019
March 2000
Controlling Woody Vegetation Page 29 Clark and Wilson
SUMMARY
Wetland prairies of the Willamette Valley, among the rarest of Oregon’s ecosystems, are threatened by invasion of woody species and non-native pest species. Because fire has been important in maintaining Willamette Valley prairies for at least 1000 years, prescribed burning is a top choice of managers for preventing encroachment of woody species. However, the effects of prescribed burning on present day wetland prairie communities, with their mix of native and non-native species, are not clear. Moreover, because strict governmental requirements for smoke management hinder prescribed burning, natural-area managers are considering alternatives to fire, like mowing and hand removal of woody plants. The present study explored the ecological interactions of a wetland prairie by experimentally testing the responses of burning, hand-removal of woody species, and mowing with removal of cut material on woody plants and key native and non-native grasses and forbs. We evaluated these perturbations for their effectiveness in meeting management objectives of reducing abundance of woody species, reducing or preventing an increase in abundance of non- native pest species, and increasing or at least maintaining native species’ abundance. After three treatments periods (1994, 1996, and 1998) the following patterns emerged: • Woody species Burning and hand-removal consistently caused the greatest reduction in cover of woody species. Mowing with removal of cut material did not reduce the abundance of woody species compared to the control. As woody plant cover decreased, plant mortality increased, indicating that treatments influenced woody plant cover at least partially through changes in survival. • Native herbaceous species All management treatments consistently failed to significantly increase the cover of Deschampsia cespitosa, while burning decreased the number of inflorescences. As a group, native graminoid flowering consistently showed no significant responses to the management treatments. Although burning and hand-removal significantly increased the cover of native forbs as group after the second treatment period in 1997, native forbs showed no significant responses to treatments in 1999. • Non-native herbaceous species Response of non-native forb cover to management treatments has been inconsistent during the five year study period with no significant
Controlling Woody Vegetation Page 29 Clark and Wilson
differences in 1995, significant reductions by burning and hand-removal in 1997, and increases with mowing in 1997. Mowing consistently increased flowering of non- native graminoids as a group. Overall no treatment was clearly superior in fulfilling restoration objectives. Burning was effective in reducing woody cover and did not promote non-native herbaceous, but it reduced the flowering the dominant native grass, Deschampsia cespitosa. Hand-removal woody species was also effective at reducing woody cover and sometimes promoted the abundance of some native species at different times, but it also at times promoted non-native herbaceous species. Because mowing with removal of cut material was ineffective in reducing woody cover and tended to promote non-native herbaceous species, this treatment is not a recommended management strategy.
INTRODUCTION
Native wetland prairies once dominated the landscape of the Willamette Valley of western Oregon (Habeck 1961, Johannessen et al. 1971). Before the mid-1800’s the Kalapuya, indigenous people of the Valley, maintained these wetland prairies by setting periodic fires during the dry season, facilitating hunting and gathering of food (Boyd 1986, Boag 1992). These fires reduced the abundance of shrubs and trees, favored the abundance of native grasses such as tufted hairgrass (Deschampsia cespitosa (L.) Beauv.), and promoted a rich variety of native forbs. Today, however, less than 1% of native wetland prairies remain (Christy and Alverson
1994). These prairies are considered among the rarest of Oregon’s ecosystems and are in critical need of conservation (ONHP 1983, Noss et al. 1995). Moreover, these remnants are refuges for many rare plant and animal species (Clark et al. 1993, Wilson et al. 1993, Wilson 1998).
Factors that contributed to the destruction of these unique wetland prairies include urban and agricultural development, threats that are also prominent today (Johannessen et al. 1971,
Towle 1982, Hammond and Wilson 1993). When settlers in the early 1800’s suppressed the large-scale fires by the Kalapuya, woody plants, such ash (Fraxinus latifolia Benth.) increased in abundance, eliminating suitable growing conditions for the prairie species (Habeck 1961,
Johannessen et al. 1971, Towle 1982). Succession to shrublands and forests now threatens most of the few remnants of undeveloped prairie (Thilenius 1968, Johannessen et al. 1971, Franklin and Dyrness 1973, Frenkel and Heinitz 1987). In addition, weedy non-native species introduced with settlement have spread rapidly with habitat disturbance, often dominating and suppressing native vegetation (Nelson 1919, Wilson and Clark, submitted).
This encroachment of woody species and invasion of non-native herbaceous pest plants are of great concern to agencies and managers responsible for conserving these remnant prairies. Controlling Woody Vegetation Page 29 Clark and Wilson
Prescribed fire is a favored choice for management because of its historical role in the Valley
(Johannessen et al. 1971, Towle 1982, Boyd 1986). Frequent burning for at least 1000 years has shaped the vegetation of these wetland prairies and returning periodic fire to the ecosystem is thought to aid in habitat restoration. Several concerns suggest caution before widespread adoption of fire as a prairie management tool. First, information on the effects of fire on wetlands is sparse, particularly in the western states, and lacks a comprehensive synthesis in contrast to the abundance of information on effects of fire in grasslands, shrublands, and forests
(Kirby et al. 1988, Schmalzer and Hinkle 1992, Kirkman 1995). Second recent invasion of prairies by non-native plants has changed the context for fire. In fact, prescribed burning might promote seedling establishment of weedy, non-native plants as much or more than the native plants (Magee 1986, Streatfeild and Frenkel 1997, Maret and Wilson, in press). Third, strict governmental smoke management rules and variable weather conditions regulate when prescribed burning can take place, which may or may not coincide with historical timing of burning. Many remnant wetland prairies are now part of urban environments surrounded by roads, businesses, and homes, and thus, smoke, although temporary, becomes a significant concern for traffic and nearby residents (Agee 1996).
Because of these concerns, studies are urgently needed to understand the ecological interactions of these unique wetlands and to evaluate the efficacy of alternate management manipulations in controlling woody species and to evaluate the effect of these manipulations, including fire, on current native and non-native herbaceous vegetation. The goal of the present study was to determine, using an experimental approach, the effects of four management alternatives, prescribed burning, mowing with removal of cut material, hand-removal of woody species, and no manipulations, on wetland prairie vegetation. We used measurements of both
Controlling Woody Vegetation Page 2 Clark and Wilson
pre-manipulation conditions and post-manipulation conditions in the analysis our data. The experimental design for the overall project (a replicated before-after-control-intervention design) gains its power from these pre-manipulation measurements.
METHODS
Study Site
Although the original vegetation of Willamette Valley wetland prairies remains somewhat unclear (Wilson et al. 1993, Streatfeild and Frenkel 1997), these seasonally flooded areas were probably dominated by bunchgrasses, particularly Deschampsia cespitosa (Savonen
1988). Other typical species may have included such graminoids as Danthonia californica
Boland, Beckmannia syzigachne (Steud.), Agrostis exarata Trin., Alopecurus aequalis Sobol.,
Eragrostis hypnoides (Lam.) B.S.P., and Juncus spp. (Nelson 1919, Habeck 1961, Moir and
Mika 1972). Today wetland prairie vegetation is characterized by a matrix of dominant grasses and sedges, interspersed with generally smaller-statured forbs (Wilson 1998).
This study was conducted on a remnant wetland prairie site (5 ha), located in west
Eugene, Oregon (123° 10’ 28”W and 44° 03’ 05”), surrounded on three sides by heavily used city streets. The site contains several types of vegetation, including upland prairie vegetation dominated by the non-native species Scot's broom (Cytisis scoparius (L.) Link) and blackberry
(Rubus discolor Weihe & Nees) and wetland prairie dominated by the native species tufted hairgrass (Deschampsia cespitosa) (Figure 1). Several Carex species, Phalaris aquatica L., and
Fraxinus latifolia dominate the wettest areas. (Figure 1). Two species on BLM's sensitive list are also present: Microcala quadrangularis and Sidalcea cusickii ssp. purpurea. A complete species list is found in Table 1.
Controlling Woody Vegetation Page 3 Clark and Wilson
The experimental plots were laid out in a section of high-quality wetland prairie dominated by Deschampsia cespitosa, but presently invaded by the woody species Fraxinus latifolia, Rosa eglanteria L., Rosa nutkana Presl, Rubus discolor, and Crataegus douglasii Lindl.
Experimental design and treatments
We used a replicated before-after-control-intervention approach with a randomized complete block design. Each of the five blocks was 22m ´ approximately 17m long, and contained four treatments: prescribed burning, mowing with removal of cut material, hand- removal of woody species, and no manipulations (Figure 2). The burn treatment areas were 8m wide, including a 1m buffer on each side. The purpose of the buffer was to reduce problems with edge effects. An additional mowed buffer (2m wide) was placed around the burn treatment for safety reasons. The other treatment areas were 7m wide, including a 1.5m buffer on each side. The length of the treatment area varied between 8m and 20m. Herbaceous plant responses were measured in three quadrats (1.0m × 0.5m) located randomly within each treatment area.
Only those quadrats containing Deschampsia cespitosa were selected. Woody plant responses were measured either within the entire treatment area or in one-half of the treatment area, depending on density of woody cover.
All treatments were applied in late summer or early fall of 1994, 1996, and 1998.
Mowing was accomplished with a hand-held weed cutter. The mowed material was gently raked off the treatment area. Hand-removal of woody species was done using hand-held equipment such as pruners and lopers cutting the woody plants at ground level.
Measurements
Controlling Woody Vegetation Page 4 Clark and Wilson
Pre-manipulation data were collected June 17-July 2, 1994. Post-manipulation data were collected June 16-22, 1995, June 23-27, 1997, and June 22-24, 1999 when the community was at the approximately the same phenological stage as during the 1994 measurements.
The cover of shrubs and trees small enough to mow with a heavy duty weed-eater was measured within a treatment area. If the density of woody species was very high, the treatment area was subsampled by dividing the area into two cells, approximately 10m long. One cell was selected randomly for the cover measurement. This cell location was applied consistently to all treatments within a block. The cover of the woody species was measured as a group and not by individual species. Woody species included shrubs Amelanchier alnifolia Nutt., Crataegus douglasii Lindl., Crataegus monogyna Jacq., Rubus discolor, Rubus laciniatus Willd., Rosa eglanteria, and Rosa nutkana and the tree Fraxinus latifolia. In all estimates of cover, two investigators reached a consensus, using calibrated templates as standards.
Survival of woody species was determined by randomly selecting three shrubs in each treatment area, tagging the plants, and mapping their location in fall of 1994. Only shrubs that were isolated enough so that individuals could be distinguished were included in the random selection. Shrub species tagged included Crataegus douglasii, Rosa spp., Rubus discolor, Rubus laciniatus, and Amelanchier alnifolia. Three Oregon ash seedlings Fraxinus latifolia were also randomly selected in each treatment area, tagged and the location mapped. The survival of these tagged individuals (< 1 m tall) was recorded on post-manipulation sampling dates in 1995 only.
The herbaceous study species, all perennials, were selected because they were either rare species of interest or as a group they accounted for most of the herbaceous cover and could indicate the response of native and non-native species to the experimental manipulations.
Nomenclature follows Hitchcock and Cronquist (1973).
Controlling Woody Vegetation Page 5 Clark and Wilson
Total herbaceous cover was measured within each quadrat along with the cover of each of the following species:
• Native forbs Sidalcea cusickii Piper ssp. purpurea, Eriophyllum lanatum (Pursh) Forbes., Veronica scutellata L., Lotus purshiana (Benth.) Clements & Clements • Non-native forbs Hypericum perforatum L, Centaurium umbellatum Gilib, Chrysanthemum leucanthemum L., Lysimachia nummularia L., Vicia sativa L., and Vicia tetrasperma L. • Native grass Deschampsia cespitosa
The number of inflorescences of each of the following species was measured in each quadrat:
• Native forbs Sisyrinchium angustifolia Mill. • Native graminoids Deschampsia cespitosa, Carex unilateralis Mack, Carex densa Bailey, Carex feta/leporina, Carex sp. 1, Juncus tenuis Willd. • Non-native grasses Holcus lanatus L. and Anthoxanthum odoratum L.
Because quadrats without Deschampsia cespitosa were not measured, the cover values reported pertain only to the permanent quadrats and do not represent the treatment areas as a whole. That is, the sampling design aims to detect species changes within permanent quadrats caused by treatments.
Data Analysis
Treatment effects were calculated as proportional changes: the change in abundance with the manipulation, divided by the initial abundance. Using proportional changes as the response variable accounts for differences among areas in pre-manipulation abundance of the measured species. To avoid problems with zeroes in the denominator, cover data that were initially zero were transformed by adding 0.1 to the zero value before calculating proportional changes. The
Controlling Woody Vegetation Page 6 Clark and Wilson
effects of treatments between 1994 and 1995 were then tested using analysis of variance
(ANOVA) when the data met the assumptions of normality and constant variance. When
ANOVA assumptions were not met, the Freidman’s Rank Sum test (SAS Institute statistical
software, version 6.08) was used. For effects of treatments between 1994 and 1999 the
Freidman’s Rank Sum test was used for all analyses. Positive values show increases in
abundances after the manipulation and negative values show declines. The significance level (a)
was set to 0.10. If main effects were significant, differences between means were tested using
Fisher’s protected least significance difference.
RESULTS
Pre-manipulation community composition
Before treatment manipulations in fall 1994, woody plants were a dominant component
of the study site with their average cover ranging between 7 and 21% within the treatment areas
(Table 2). Of the herbaceous species measured, Deschampsia cespitosa contributed the most cover, ranging between 15 and 22%. The average cover for the measured group of common native forb species ranged between 2 and 4%. Lotus purshiana was the most abundant native forb. The average number of inflorescences for the measured group of common native graminoid species was about 10 to 13, with Deschampsia cespitosa producing the highest average (Table 2).
The average cover for the measured group of common non-native forb species ranged between 1 and 6% (Table 2). Hypericum perforatum and Vicia tetrasperma were the most consistently abundant non-native forbs. Lysimachia nummularia had the highest abundance of any forb within a single treatment area, but was found only in block 2. The average number of
Controlling Woody Vegetation Page 7 Clark and Wilson
inflorescences for the measured group of common non-native graminoid species was about 5 to
14. Anthoxanthum odoratum and Holcus lanatus both had similar average number of inflorescences. However, Anthoxanthum odoratum was much more variable, reflecting its patchy distribution on the site.
Average overall herbaceous cover within quadrats ranged between 35 and 39% (Table 2).
Deschampsia cespitosa cover accounted for approximately half of this overall cover. Because quadrats without Deschampsia cespitosa were not measured, the cover values reported in Table 2 pertain only to the permanent quadrats and do not represent the treatment areas as a whole.
Effects of experimental manipulations
Woody species. Burning and hand removal significantly decreased the cover of woody species more than the decrease found in control plots the first season after treatments in fall 1994 (Table
3). The 44% decrease caused by mowing, however, was similar to the 32% decrease found in the control. In 1997 after repeating treatments in fall 1996, the pattern of response was similar to that of the initial response in 1995 with burning and hand-removal significantly reducing cover of woody species (-63% and –79%) compared to the control (+20%) (Table 3). Woody cover after mowing with removal of cut material (+25%) was not significantly different from the increase in cover in the control (+20%).
After repeating the treatments a third time in fall 1998, the responses of woody plants in
1999 followed the previous patterns with burning and hand-removal significantly reducing woody cover (81% and 65%) compared to the control (144%) (Table 3). As in previous years, the change in woody cover after mowing with removal of cut material (56%) was not significantly different from that in the controls (+144%) (Table 3). The responses of woody
Controlling Woody Vegetation Page 8 Clark and Wilson
cover to burning and hand-removal remained similar between years, whereas woody cover initially decreased with mowing in 1995 and then increased after the remaining two treatment periods similar to the patterns with the control (Table 3).
Survival of woody species, measured only the first year after treatment, showed that significantly fewer shrubs survived burning (+33%) compared to the control (+88.3%) (Table 3).
In contrast, survival with hand removal (+60.0%) and mowing (+83.3%) were not significantly different from the control.
Native herbaceous species. None of the treatments had a significant effect on the cover of
Deschampsia cespitosa after any of the three treatment periods, although burning consistently produced the highest values (Table 4). Burning, however, reduced the relative number of
Deschampsia cespitosa inflorescences in each of the treatment periods (+192%, +99%, +1%) compared to the controls (+246%, +312%, +65%), although the difference was only statistically significant in 1997 (Table 4). In contrast, burning and mowing increased the inflorescence number of Juncus tenuis in 1995 (+4374%, 1625%), in 1997 (+474%, +608%), and in 1999
(+1314%, +1238%) compared to the control (+573%,+93%, +260%) with significant increases only in 1997 (Table 4) As group, however, native graminoid species showed no significant differences in flowering response to the treatments during any of the three treatment periods
(Table 4).
In 1997, burning and hand removal significantly promoted the cover of native forbs as a group in 1997 (+8% and +167%) (Table 4) compared to the controls (-77%). The group response to mowing (-14%) was similar to that of the controls. The two species chiefly contributing to this group increase in 1997 were Lotus purshiana, with burning (- 4%) and hand
Controlling Woody Vegetation Page 9 Clark and Wilson
removal (+1483%) relatively promoting cover compared to the controls (-64%) and Veronica scutellata, with hand removal (+79%) and mowing (+98%) significantly increasing cover (Table
4). In 1999, however, there were no significant differences between treatments for this two species (Table 4). Similarly, the cover of native forbs as a group was not significantly different between treatments in 1999 (Table 4).
The native forb Sisyrinchium idahoensis flowered for the first time since the study began in 1997 in the mow treatment and flowered for the first time in the burn treatment in 1999 (Table
4). The lack of proportional change in number of inflorescences of Sisyrinchium idahoensis in previous years was due to its absence in the sampling plots. A second native forb Eriophyllum lanatus, which was previously absent in the burn treatments, also appeared for the first time in a burn treatment in 1999 (Table 4).
Non-native herbaceous species. In the first season after treatments (1995), burning, hand- removal, and mowing ameliorated the decline of cover in the controls of two non-native herbaceous species, Vicia tetrasperma and Vicia sativa (Table 4). This response pattern disappeared by 1997 for Vicia tetrasperma, but remained for Vicia sativa through 1999 (Table
4). In 1997, burning and hand-removal significantly reduced cover of Hypericum perforatum (-
89% and -78%) (Table 4), compared to the cover in the controls (+92%). In the following year, however, there were no significant differences between treatments (Table 4).
Although burning reduced the cover of Chrysanthemum leucanthemum, this species was growing, just by chance, only in the burn treatments in 1995 and 1997. In 1999, however,
Chrysanthemum leucanthemum was growing for the first time in the control plots. The 1999
Controlling Woody Vegetation Page 10 Clark and Wilson
cover of Chrysanthemum leucanthemum increased (3%) in the burn treatment compared to that of previous years (-35% and -28%) (Table 4). Cover consisted mostly of seedlings in these plots.
The response of non-native species cover as a group has been inconsistent among years with no significant differences in 1995, significant reductions by burning (-30%) and hand- removal (-45 %) compared to the control (+28) in 1997, and increases in cover with mowing
(+232%) and reduction with burning (-44%) compared to the control (-15%) in 1999 (Table 4).
All three treatments increased the number of inflorescences of Anthoxanthum odoratum after the first two treatment periods compared to the declines in the control, although only the
1997 burn and mow treatments were statistically different from the controls (Table 4). This pattern was changed in 1999 with reductions of flowering after burning and hand-removal
(139%, + 80%) compared to the control (+322%) (Table 4). Hand removal and mowing increased the number of inflorescences of Holcus lanatus for all three treatment periods compared to declines in both the burned plots and controls though the differences were statistically different only in 1997 (Table 4). The burn significantly decreased the inflorescence number (-90%) of Holcus lanatus relative to the the decrease of the control (-19%) in 1999
(Table 4). As a group, mowing significantly promoted flowering of the graminoid species
(+147%) compared to the decrease in the controls (-18%) (Table 4).
Total cover. All three treatments increased the total cover in 1995, with mowing causing a significant increase (+32%) (Table 4). In 1997, however, there was little difference in total cover between treatments (Table 4). In 1999, all three treatments again increased total cover, with both mowing and hand-removal causing significant increases (72%, 42%) compared to the control
Controlling Woody Vegetation Page 11 Clark and Wilson
(-9%). The total cover is for all species present in a quadrat, not the total of the measured
species.
DISCUSSION
Effects of experimental manipulations
Patterns and mechanisms of woody species responses. Prescribed burning and hand removal of woody species consistently reduced the overall cover of woody species to very low levels. In contrast, the cover of woody species in mowed plots did not differ significantly from that in unmanipulated controls. These differences in treatment effects on overall cover can be understood in terms of processes at the individual level. All the woody species in this study--
Amelanchier alnifolia, Crataegus douglasii, Crataegus monogyna, Rubus discolor, Rubus laciniatus, Rosa eglanteria, Rosa nutkana, and Fraxinus latifolia -- can resprout after the removal of their biomass (Tiermenstein 1989, Reed 1993, Pendergrass 1996, Clark and Wilson, personal observation), and hence have the potential for quickly regaining cover. Thus, lasting reduction of cover of these species requires killing individual plants. In this study, differences among treatments in the reduction in woody cover closely followed mortality rates (Table 3).
The lowest survival was in burned plots, which also had significant reductions in woody plant cover. Survival did not vary between mowed and control plots, and neither did change in woody plant cover. The strong reduction in cover with hand removal coupled with only moderate mortality possibly occurred because clipping of stems close to the ground removed more basal vegetative buds as compared to mowing and so reduced the capacity of clipped plants to resprout.
Controlling Woody Vegetation Page 12 Clark and Wilson
Fire is effective in reducing the abundance of invading woody species in drier prairie environments (Daubenmire 1968, Vogl 1974, Kozlowski and Ahlgren 1974, Bragg and Hulbert
1976, Kucera 1981, Collins and Wallace 1990). Few studies, however, investigate the effects of fire on wetland vegetation, and in particular, prairie wetlands (Kirby et al. 1988, Schmalzer and
Hinkle 1992, Kirkman and Sharitz 1994, Kirkman 1995). The handful of studies investigating fire effects on Willamette wetland prairie vegetation show inconsistent responses of woody species (Streatfeild 1995, Clark and Wilson 1998, Acker 1991, Pendergrass 1996). Differences in the patterns of woody plant responses may be due to variation in fire characteristics of the prescribed burn and to variation in the capacity of woody species to resprout after fire.
Moreover, the way woody response is measured (density, cover, height, and biomass) can also confound general patterns.
Patterns and mechanisms of herbaceous species responses. The three management treatments, prescribed burning, mowing, and hand-removal, failed to significantly increase the cover and number of inflorescences of Deschampsia cespitosa with burning generally reducing the number of inflorescences. Other studies in Willamette Valley wetland prairies show contrasting responses of Deschampsia cespitosa to burning. For example, at the Finley Wildlife Refuge burning decreased aboveground vegetative and reproductive biomass and inflorescence number of Deschampsia cespitosa (Clark and Wilson 1998). Recently burned areas at the same site had reduced cover compared to earlier burned areas (Streatfeild and Frenkel 1997). In other wetland prairies in west Eugene, Deschampsia cespitosa frequency increased immediately after burning, but cover declined significantly, suggesting a shift from fewer large plants to a greater number of smaller plants (Pendergrass 1996). Cover, however, returned to preburn abundance the following year (Pendergrass 1996). A similar pattern was observed in Horkelia Prairie (west
Controlling Woody Vegetation Page 13 Clark and Wilson
Eugene) where Deschampsia cespitosa cover decreased almost 50% the first year after burning and then increased to slightly more than in the controls two years after the burn (Wilson et al.
1995).
For the other graminoid species, Juncus tenuis (native), Anthoxanthum odoratum (non- native) and Holcus lanatus (non-native), mowing and hand-removal generally promoted flowering. Both hand-removal and mowing reduced the height of woody plants (even though mowing did not reduce woody cover), presumably releasing the graminoids from competitive suppression by shading. The mowing treatment also reduced the biomass of cut plants and removed the cut material from the plots, which can promote herbaceous growth (Daubenmire
1968).
Burning of these graminoid species produced mixed results , increasing flowering of
Juncus tenuis for all three treatment periods and of Anthoxanthum odoratum for only the first two treatment periods. Burning decreased the number of inflorescences of Holcus lanatus in the last treatment period. Increases in flowering by graminoids and other monocotyledons after fire in grasslands have often been observed (Daubenmire 1968, Gill 1981, Whelan 1995). Burning, although it can lead to competitive release with a decrease in woody species cover, might not have increased flowering in Holcus lanatus because of direct mortality of its tillers (Grime and
Lloyd 1973, Wilson and Clark 1997, but see Pendergrass 1996, Clark and Wilson 1998).
Patterns of forb responses to repeated management treatments have generally been inconsistent for individual species and for groups of native and non-native species during the entire study period. Due to lag time in response to disturbance, patterns of vegetation responses to repeated treatments may take several years to develop (Wilson and Clark, submitted). Five years after the initiation of this study, two native forbs Sisyrinchium idahoensis and Eriophyllum
Controlling Woody Vegetation Page 14 Clark and Wilson
lanatus appeared for the first time in the burn treatments. Response to disturbances may be modified by interactions with other environmental factors. Management treatments had little effect on the response of Veronica scutellata in both 1995 and 1999, but in 1997 all treatments increased cover relative to the control. Fires differ in their characteristics from year to year, potentially altering vegetation responses. For example, Veronica scutellata established or significantly increased in frequency in burned areas of west Eugene wetland prairies
(Pendergrass 1996). In contrast, fire was detrimental to Veronica scutellata at the Finley
Wildlife Refuge, with significant decreases in survival, aboveground vegetative and reproductive biomass, and number of inflorescences compared to unburned treatments (Clark and Wilson
1998). Also in another study at the Finley wetland cover was lower in more recently burned areas compared to areas burned earlier (Streatfeild and Frenkel 1997).
Management Implications
Management strategies for restoration of Willamette Valley native prairies require reduction in the abundance of woody species, the reduction or prevention of an increase in abundance of non-native pest plant species, and the increase or at least the maintenance of native plant species abundance (Table 5). After three treatment periods (1994, 1996, 1998), both prescribed burning and hand-removal of woody species reduced the abundance of woody species, indicating that hand-removal is a viable alternative to prescribed burning (Table 5).
Given, however, the ability of these woody species to resprout after disturbance, long-term reduction of woody species abundance requires death of the plant by repeated treatments. The pattern established thus far indicates that woody cover reduction is maintained with two-year treatment intervals of burning and hand-removal. In contrast to the other two treatments,
Controlling Woody Vegetation Page 15 Clark and Wilson
mowing with removal of cut material did not reduce the abundance of woody species and thus, would not be a viable alternative management option (Table 5).
Because fire was used historically in the past to maintain native prairie of the Willamette
Valley, managers predict that prescribed burning will promote or maintain the abundance and flowering of fire-adapted native herbaceous species and that fire will reduce or at least not increase the abundance and flowering of non-native herbaceous species, which may not be adapted to frequent fire. The results of this investigation show that native herbaceous species as a group did not change in abundance and flowering compared to the controls, supporting predictions. One important exception was the reduction by fire of flowering of the key native grass, Deschampsia cespitosa. The non-native herbaceous species as a group did not change in cover or flowering compared to the controls also supporting predictions (Table 5).
The alternative management options to burning, hand removal and mowing, were generally successful in meeting the management objectives for herbaceous species (Table 5), with the exception of both options producing an unwanted increase in the flowering of non- native graminoid species in 1995 and again with mowing in 1999.
In summary, the results to date show that burning and hand-removal of woody species are the best options of the three management treatments for achieving the restoration goals for native wetland prairies (Table 5). Because of the ineffectiveness of mowing with removal of cut material in decreasing cover of woody species combined with the tendency to promote flowering of non-native herbaceous species, we do not recommend mowing as management strategy.
Consideration of other criteria, such as economics, availability of fire-fighters, and suitability of weather conditions for burning, will be necessary for choosing between burning and hand removal.
Controlling Woody Vegetation Page 16 Clark and Wilson
Even though this study is one of the very few long-term investigations of effects of management manipulations on wetland prairies in the Willamette Valley, only repeated treatments and continued careful measurements will demonstrate conclusively the value of these management approaches.
ACKNOWLEDGMENTS
We gratefully acknowledge the financial and logistical support of the Bureau of Land
Management, who is responsible for the management of Danebo wetland. A Challenge Cost
Share agreement between the Bureau of Land Management, Coast Range Area, Eugene, OR and
Oregon State University, Corvallis, OR provided the financial support for this project. The experimental treatments were conducted by the Bureau of Land Management, supervised by
Kathy Pendergrass, Bureau of Land Management botanist.
Many other people contributed to the success of this project. We thank Amy Clark, Tina
Dreisbach, Linda Hasselbach, Keli Kuykendall, Scott Wadlow, Machelle Nelson, and Marilynn
Bartels for their assistance in the field. We also thank Elida Machado, Scott Wadlow, and
Mahala Young for help in data entry.
REFERENCES
Acker, S.A. 1991. Analysis of experimental tree burning at the Willow Creek Preserve, Lane
County, Oregon. Unpublished report on file at the Oregon Field Office of The Nature
Conservancy, USA.
Controlling Woody Vegetation Page 17 Clark and Wilson
Agee, J.K. 1996. Achieving conservation biology objectives with fire in the Pacific Northwest.
Weed Technology 10:417-421.
Boag, P.G. 1992. Environment and Experience: Settlement Culture in Nineteenth-Century
Oregon. University of California Press, Berkeley, CA, USA.
Boyd, R. 1986. Strategies of Indian burning in the Willamette Valley. Canadian Journal of
Anthropology 5:65-86.
Christy, J.A. and E.R. Alverson. 1994. Saving the Valley’s wet prairie. The Nature
Conservancy –Oregon Chapter Newsletter. Portland, Oregon, USA.
Clark, D.L. and M.V. Wilson. 1998. Fire effects on wetland prairie plant species. Report to the
U.S. Fish and Wildlife Service, Western Oregon Refuges, Corvallis, Oregon, USA.
Clark, D.L., K.K. Finley, C.A. Ingersoll. 1993. Status Report for Erigeron decumbens var. decumbens. Report prepared for the Conservation Biology Program, Oregon Department of
Agriculture, Salem, Oregon, USA
Collins, S.L. and L.L. Wallace (eds) 1990. Fire in North American Tallgrass Prairies.
University of Oklahoma Press: Norman, OK, USA.
Controlling Woody Vegetation Page 18 Clark and Wilson
Daubenmire, R.F. 1968. Ecology of Fires in Grasslands. IN Advances in Ecological Research
Vol. X. J.B. Cragg (ed) p. 209-266. Academic Press. New York, NY, USA.
Finley, K.K. 1995. Hydrology and related soil features of three Willamette Valley wetland prairies. M.S. Thesis. Oregon State University, Corvallis, OR, USA.
Franklin, J. F. and C.T. Dyrness. 1973. Natural vegetation of Oregon and Washington. USDA
Forest Service General Technical Report. PNW-8, Pacific Northwest Forest Range Experiment
Station. Portland, Oregon, USA.
Frenkel, R.E. and E.R. Heinitz. 1987. Composition and structure of Oregon ash (Fraxinus latifolia) forest in William L. Finley National Wildlife Refuge, Oregon. Northwest Science
61:203-212.
Gill, A.M. 1981. Fire adaptive traits of vascular plants. IN H.A. Mooney, T.M. Bonnicksen,
N.L. Christensen, J.E. Lotan, and W.A. Reiners (eds). Fire Regimes and Ecosystem Properties,
US Forest Service, General Technical Report WO-26, Washington, D.C., USA.
Grime, J.P. and P.S. Lloyd. 1973. An Ecological Atlas of Grassland Plants. Fletcher and Son
LTD, Norwich.
Habeck, R. J. 1961. The original vegetation of the mid-Willamette Valley. Northwest Science
35:65-77.
Controlling Woody Vegetation Page 19 Clark and Wilson
Hammond, P.C. and M.V. Wilson. 1993. Status of the Fender’s blue butterfly. Report to the
U.S. Fish and Wildlife Service, USA.
Hitchcock, C.L. and A. Cronquist. 1973. Flora of the Pacific Northwest. University of
Washington Press, Seattle, USA.
Holttum, R.E. 1955. Growth habits of monocotyledons -- variations on a theme.
Phytomorphology 5:399-413.
Huffaker, C. B. 1964. Fundamentals of biological weed control IN (DeBach, P. and Schlinger,
E.I. (eds). Biological Control of Insect Pests and Weeds, Chapman and Hall, London, pages 74-
117.
Johannessen, C.L., W.A. Davenport, A. Millet, and S. McWilliams. 1971. The vegetation of the
Willamette Valley. Annals of the Association of American Geographers 61:286-302.
Kirkman, L.K. 1995. Impacts of fire and hydrological regimes on vegetation in depression wetlands of southeastern USA. In Proceedings 19ths Tall Timbers Fire Ecology Conference Fire in wetlands: A Management Perspective. Tall Timbers Research Station, Tallahassee, Florida.
Kirkman, L.K. and R.R. Sharitz. 1994. Vegetation disturbance and maintenance of diversity in intermittently flooded Carolina Bays in South Carolina. Ecological Applications 4:177-188.
Controlling Woody Vegetation Page 20 Clark and Wilson
Kirby, R.E., S.J. Lewis, and T.N. Sexson. 1988. Fire in North American Wetland Ecosystems and Fire-Wildlife Relations: An Annotated Bibliography. Fish and Wildlife Service, U.S.
Department of the Interior, Washington, D.C., USA.
Kozlowski, T.T. and C.E. Ahlgren. 1974. Fire and Ecosystems. Academic Press, New York,
USA.
Kucera, C.L. 1981. Grasslands and fire. P. 90-111. In H.A. Mooney, T.M. Bonnicksen, N.L.
Christensen, J.E. Lotan, and W.A. Reiners (eds). Fire Regimes and Ecosystem Properties, US
Forest Service, General Technical Report WO-26, Washington, D.C., USA.
Lange, O.L., P.S. Nobel, C.B. Osmond, H. Ziegler. 1981. Physiological Plant Ecology I,
Springer-Verlag, Berlin, New York, USA.
Lugo, A.E. 1995. Fire and Wetland Management. In Proceedings 19ths Tall Timbers Fire
Ecology Conference Fire in wetlands: A Management Perspective. Tall Timbers Research
Station, Tallahassee, Florida, USA.
Magee, T. K. 1986. Vegetation monitoring at Wren Prairie Preserve: I. Initial response of grassland vegetation to controlled burning. Unpublished report on file at the Oregon Field
Office of The Nature Conservancy, Portland, Oregon. USA.
Controlling Woody Vegetation Page 21 Clark and Wilson
Maret, M.P. and M.V. Wilson. (in press). Fire and seedling dynamics in western Oregon prairies. Journal of Vegetation Science.
Michener, W.K. 1997. Quantitatively evaluating restoration experiments: research design, statistical analysis, and data management considerations. Restoration Ecology 5:324-337.
Moir, W. and P. Mika. 1972. Prairie vegetation of the Willamette Valley, Benton County,
Oregon. Report on file at Forest Science lab, 3200 Jefferson Way, Corvallis, Oregon, USA.
Montalvo, A.M., S.L. Williams, K.J. Rice, S.L. Buchmann, C. Cory, S.N. Handel, G.P. Nabhan,
R. Primack, R.H. Robichaux. 1997. Restoration biology: A population biology perspective.
Restoration Ecology 5: 277-290.
Nelson, J.C. 1919. The grasses of Salem, Oregon, and vicinity. Torreya 19:216-227.
Noss, R.F., E.T. LaRoe III, J.M. Scott. 1995. Endangered ecosystems of the United States: A preliminary assessment of loss and degradation. National Biological Service, Biological Report
28.
Oregon Natural Heritage Program. 1983. Survey of Willamette Valley 1981-1983 (A report to the Mason Trust). The Natural Conservancy. Portland, Oregon, USA.
Controlling Woody Vegetation Page 22 Clark and Wilson
Pickett, S.T.A. and V.T. Parker. Avoiding the old pitfalls: opportunities in a new discipline.
Restoration Ecology 2:75-79.
Palmer, M.A., R.F. Ambrose, N.L. Poff. 1997. Ecological theory and community restoration ecology. Restoration Ecology 5: 291-300.
Pendergrass (Connelly), K.L. 1996. Vegetation composition and response to fire of native
Willamette Valley wetland prairies. M.S. Thesis, Oregon State University, Corvallis, Oregon.
USA.
Reed, William R. 1993. Rosa nutkana. In: Fischer, William C., compiler. The Fire Effects
Information System [Data Base]. Missoula, MT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station, Intermountain Fire Sciences Laboratory. Magnetic tape reels; 9 track; 1600 bpi, ASCII with Common LISP present.
Savonen, C. 1988. Historical wetlands of the West Eugene Study Area. Unpublished report on file at the Oregon Field Office of The Nature Conservancy, USA.
Schaffers, A.P., M.C. Vesseur, and K.V. Sykora. 1998. Effects of delayed hay removal on the nutrient balance of roadside plant communities. Journal of Applied Ecology 35: 349-364.
Schmalzer, P.A. and C.R. Hinkle. 1992. Soil dynamics following fire in Juncus and Spartina marshes. Wetlands 12:8-21.
Controlling Woody Vegetation Page 23 Clark and Wilson
Soil Conservation Service. 1975. Soil survey of Benton County Area, Oregon. U.S.
Department of Agriculture.
Streatfeild, R. 1995. Ecological survey and interpretation of the Willamette Floodplain Research
Natural Area. W.L. Finley National Wildlife Refuge, Oregon. M.S. Thesis, Oregon State
University. Corvallis, OR, USA.
Streatfeild, R. and R.E. Frenkel. 1997. Ecological survey and interpretation of the Willamette
Floodplain Research Natural Area, W. L. Finley National Wildlife Refuge, Oregon, USA.
Natural Areas Journal 17:346-354.
Tiermenstein, D. 1989. Rubus discolor. In: Fischer, William C., compiler. The Fire Effects
Information System [Data Base]. Missoula, MT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station, Intermountain Fire Sciences Laboratory. Magnetic tape reels; 9 track; 1600 bpi, ASCII with Common LISP present.
Thilenius, J.F. 1968. The Quercus garryana forest of the Willamette Valley, Oregon. Ecology
49:1124-1133.
Towle, J.C. 1982. Changing geography of Willamette Valley woodlands. Oregon Historical
Quarterly 83:66-87.
Controlling Woody Vegetation Page 24 Clark and Wilson
Vogl, R. 1974. Effects of fire on grasslands. In Kozlowski, T.T. and C.E. Ahlgren (eds.). Fire and Ecosystems. Academic Press, New York. USA.
Whelan, R.J. 1995. The Ecology of Fire. Cambridge University Press.
Wilson, M.V. 1998. Wetland Prairie, Part I of the U.S. Fish and Wildlife Service Willamette
Basin Recovery Plan, prepared for U.S. Fish and Wildlife Service, Oregon State Office, USA.
Wilson, M.V. and D.L. Clark. 1997. Effects of fire, mowing, and mowing with herbicide on native prairie of Baskett Butte, Baskett Slough NWR. Final report 1994-1997 to the U.S. Fish and Wildlife Service, Western Oregon Refuges, USA.
Wilson, M.V. and D.L. Clark (submitted) Controlling invasive tall oatgrass Arrhenatherum elatius and promoting native grasses through mowing. Journal of Applied Ecology.
Wilson, M.V. and L. Lantz. (submitted) Teaming up for restoration. Ecological
Restoration/North America.
Wilson, M.V., K.P.Connelly (Pendergrass) and L. Lantz. 1993. Plant species, habitat, and site information for Fern Ridge Reservoir. A component of the Project to Develop Management
Guidelines for Native Wetland Communities. Prepared for Waterways Experiment Station,
Army Corps of Engineers, Vicksburg, Mississippi, and the Soil Conservation Service, Portland,
Oregon.
Controlling Woody Vegetation Page 25 Clark and Wilson
Wilson, M.V., C.A. Ingersoll, and D. Clark. 1995. Seed dormancy, germination, and establishment. In Studies of plant establishment limitations in wetlands of the Willamette
Valley, Oregon, Army Corps of Engineers, Waterways Experiment Station; prepared for U.S.
Army Crops of Engineers. Wetlands Research Program Technical Report WRP-RE-13), USA,
75 pages.
Wright, H. A. and A.W. Bailey. 1982. Fire Ecology – Western United States and Southern
Canada. John Wiley and Sons, New York, USA.
Controlling Woody Vegetation Page 26 Clark and Wilson
Figure 1. Map of vegetation types at the BLM Danebo wetland office site (compiled by Glenn Miller, Oregon Department of Agriculture and provided by the BLM Coast Range Office)
Controlling Woody Vegetation Page 27 Clark and Wilson
Figure 2. Map of the Danebo study site showing the 5 blocks (not drawn to scale) and the location of treatments within the blocks: M is mow with litter removal, H is hand-removal of woody species, C is no manipulation and B is burn.
Controlling Woody Vegetation Page 28 Clark and Wilson Table 1. A list of vascular plants found at the Danebo wetland prairie site. All species are in the Willamette Valley wet prairie community unless otherwise indicated. The species list was compiled by Peter Zika, September 1988, and provided by the BLM Coast Range Office. *Alien, established in the Oregon flora after white settlement; 1Upland disturbed soils; 2Planted; 3Sensitive Species
Acer macrophyllum1 Crataegus douglasii Lathyrus sphaericus* Pyrus communis*1 Acer saccharinum*2 Crataegus monogyna*1 Leontodon nudicaulis* Pyrus fusca Agrostis capillaris*1 Cynosurus echinatus* Liliaceae sp.*2 Pyrus malus*1,2 Agrostia stolonifera Dactylis glomerata*1 Lolium perenne*1 Ranunculus aquatilis Aira caryophyllea*1 Danthonia californica Lotus corniculatus*1 Ranunculus Allium amplectens Daucus carota*1 Lotus formosissimus occidentalis Allium vineale* Deschampsia cespitosa Lotus purshianus Ranunculus Alopecurus geniculatus Digitalis purpurea*1 Lysimachia orthorhynchus Alopecurus pratensis*1 Dipsacus sylvestris*1 nummularia* Rhus diversiloba Amelanchier alnifolia Downingia yina Madia glomerata Rosa eglanteria*1 Anthemis cotula*1 Eleocharis acicularis Madia sativa Rosa nutkana Anthoxanthum Eleocharis palustris Malva neglecta*1 Rosa sp. odoratum*1 Epilobium glandulosum Mentha pulegium* Rosa multiflora*1 Araliaceae sp.*1,2 Epilobium paniculatum Microcala Rubus discolor*1 Arrenatherum elatius*1 Epilobium watsonii quadrangularis3 Rubus laciniatus*1 Aster hallii Eriophyllum lanatum Microseris laciniata Rubus leucodermis1 Avena barbata*1 Eryngium petiolatum Microsteris gracilis Rumex acetosella* Barbarea orthoceras Eschscholzia Moenchia erecta* Rumex crispus* Beckmannia californica1 Montia fontana Rumex sp. syzigachne Festuca arundinacea* Montia linearis Salix lasiandra Bellis perennis*1 Festuca bromoides* Myosotis discolor* Salix piperi Berberis sp.*1,2 Festuca megalura*1 Narcissus pseudo- Salix piperi Betula pendula* Fraxinus latifolia narcissus* x sitchensis Bidens frondosa Galium aparine Oemleria cerasiformis Salix scouleriana Boisduvalia densiflora Galium parisiense* Parentucellia viscosa*1 Salix sitchensis Brassica hirta* Geranium dissectum* Phalaris aquatica* Scilla sp.*2 Briza minor* Gnaphalium palustre Phalaris arundinacea Sanguisorba minor*1 Brodiaea hyacinthina Grindelia squarossa Phleum pratense* Senecio vulgaris* Bromus carinatus Hedera helix*1,2 Picea pungens*1,2 Senecio jacobea*1 Bromus mollis*1 Holcus lanatus* Pinus sp.*1,2 Scirpus acutus Camassia quamash Hordeum Plagiobothrys figuratus Sidalcea cusickii Cardamine oligosperma brachyantherum Plagiobothrys scouleri ssp. purpurea3 Carex densa Hordeum murinum*1 Plantago lanceolata*1 Sisyrinchium Carex echinata Hypericum Poa annua*1 idahoensis Carex feta perforatum* Poa compressa*1 Spergularia rubra*1 Carex leporina Hypocharis radicata* Poa pratensis* Syringia vulgaris*1,2 Carex sp. (ovales) Ilex aquifolium*1,2 Poaceae spp. Taraxacum officinale* Carex unilateralis Iris sp.*2 Polygonum aviculare*1 Thuja plicata1,2 Centaureum Juncus bufonius Polygonum douglasii Trifolium dubium* umbellatum* Juncus effusus Populus nigra*1,2 Typha latifolia Cerastium viscosum* Juncus ensifolius Populus trichocarpa Veronica arvensis* Chrysanthemum Juncus nevadensis Potentilla sp. Veronica scutellata leucanthemum*1 Juncus oxymeris Prunella vulgaris* Vicia cracca*1 Chrysanthemum Juncus patens Prunus avium*1 Vicia hirsuta*1 parthenium*1 Juncus tenuis Prunus domestica1,2 Vicia sativa*1 Cirsium arvense*1 Lactuca serriola*1 Pseudotsuga Vicia tetrasperma* Cirsium vulgare*1 Lapsana communis*1 mensiesii1,2 Convolvulus arvensis*1 Lathyrus latifolius*1
Controlling Woody Vegetation Page 29 Clark and Wilson
Controlling Woody Vegetation Page 30 Clark and Wilson
Table 2. Average cover (%) of woody species and average cover (%) or number of inflorescences of common herbaceous plants in 1994 before management treatments. Woody species include shrubs Rubus discolor, Rubus laciniatus, Rosa nutkana, Rosa eglanteria, Crataegus douglasii, Crataegus monogyna, Amelanchier alnifolia and the tree Fraxinus latifolia. Values within parentheses are standard deviations. Treatments
Burn Hand Mow Control removal Woody species (cover) 10.5 (7.7) 21.3 (21.0) 7.1 (4.6) 12.9 (6.0)
Common native graminoid species
Cover
Deschampsia cespitosa 18.0 (6.5) 18.7 (4.4) 21.5 (3.9) 15.3 (4.0)
Number of inflorescences
Carex unilateralis 0.0 (0.0) 2.1 (4.1) 1.0 (2.2) 0.3 (0.4)
Carex densa 1.4 (2.1) 0.2 (0.4) 0.7 (1.1) 1.2 (1.2)
Carex feta/leporina 0.3 (0.4) 0.4 (0.9) 1.7 (2.2) 0.3 (0.3)
Carex sp. 1 1.4 (1.8) 0.4 (0.9) 0.9 (1.5) 1.6 (0.6)
Deschampsia cespitosa 6.3 (4.3) 6.7 (2.7) 7.9 (5.4) 4.7 (2.3)
Juncus tenuis 1.3 (0.7) 0.9 (0.8) 1.1 (0.6) 3.1 (2.3)
Group total 10.6 (5.3) 10.7 (5.6) 13.3 (5.5) 11.1 (4.5)
Common native forb species
Cover Eriophyllum lanatum 0.0 (0.0) 0.9 (1.6) 0.0 (0.0) 0.5 (1.0)
Lotus purshiana 1.5 (1.2) 1.2 (1.4) 1.7 (1.6) 2.2 (2.0)
Sidalcea cusickii ssp. purpurea 0.0 (0.0) 0.0 (0.0) 0.1 (0.2) 0.0 (0.0)
Veronica scutellata 0.8 (1.2) 0.8 (1.1) 0.6 (1.0) 1.3 (2.1)
Group total 2.3 (1.2) 2.9 (1.8) 2.4 (1.4) 4.0 (3.0)
Number of inflorescences
Sisyrinchium idahoensis 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0)
Common non-native graminoid species
Number of inflorescences
Anthoxanthum odoratum 0.9 (1.3) 2.4 (5.4) 4.0 (7.4) 10.7 (23.7)
Holcus lanatus 5.8 (3.7) 2.7 (2.5) 2.9 (3.5) 3.6 (2.1)
Controlling Woody Vegetation Page 31 Clark and Wilson
Group total 6.7 (4.6) 5.1 (4.3) 6.9 (6.2) 14.3 (23.0)
Common non-native forb species
Cover
Centaurium umbellatum 0.5 (0.6) 0.1 (0.2) 0.5 (0.7) 0.8 (0.9)
Chrysanthemum leucanthemum 0.3 (0.6) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0)
Hypericum perforatum 1.4 (1.2) 1.7 (1.5) 0.7 (1.0) 1.1 (0.5)
Lysimachia nummularia 0.0 (0.0) 2.7 (6.0) 0.0 (0.0) 3.4 (7.6)
Vicia sativa 0.0 (0.0) 0.1 (0.1) 0.0 (0.0) 0.3 (0.3)
Vicia tetrasperma 1.6 (1.0) 1.1 (1.7) 1.1 (1.0) 1.2 (0.7)
Group total 2.2 (1.0) 4.5 (5.6) 1.2 (1.6) 5.3 (7.2) Overall cover (all species) 39.1 (6.3) 35.1 (5.5) 36.2 (4.3) 38.4 (8.6)
Controlling Woody Vegetation Page 32 Clark and Wilson
Table 3. Average percent survival and average proportional change in cover (%) for woody species between 1994 and 1995, between 1994 and 1997, and between 1994 and 1999 within treatment areas. Reported values are untransformed means. Positive values of cover show increases in abundances after the manipulation; negative values show declines. P are the probabilities that survival and proportional changes differed significantly among treatments just by chance (Freidman’s Rank Sum Test). Treatment means with the same letters were statistically indistinguishable.
Treatment Burn Hand removal Mow Control P
Survival 1995 33.3b 60.0ab 88.3a 83.3a 0.03
Cover 1994-1995 -0.91bc -0.82c -0.44ab -0.32a 0.02
1994-1997 -0.63b -0.79b 0.25a 0.20a 0.00
1994-1999 -0.81b -0.65b 0.56a 1.44a 0.00
Controlling Woody Vegetation Page 33 Clark and Wilson
Table 4. Average proportional change in cover and number of inflorescences of common herbaceous species within treatment areas between 1994 and 1995, between 1994 and 1997, and between 1994 and 1999. Average proportional changes for groups of native and non-native species are also reported. Reported values are untransformed means. Positive values show increases in abundances after the manipulation; negative values show declines. P is the probability that the proportional changes differed significantly just by chance (Freidman’s Rank Sum Test, ANOVA). Treatments sharing letters were statistically indistinguishable. Species Treatment
Hand Burn removal Mow Control P
Common native graminoids Cover Deschampsia cespitosa 1995 0.25 -0.14 -0.25 -0.11 0.50 1997 0.31 0.08 -0.01 0.31 0.77 1999 0.42 0.21 0.07 -0.06 0.92 Number of inflorescences Carex unilateralis 1995 0.00 0.45 -0.08 0.47 0.56 1997 0.00 0.66 0.36 0.27 0.99 1999 1.80 -0.40 0.00 0.00 0.43
Carex densa 1995 0.74 -0.20 0.58 -0.50 0.61 1997 7.00 -0.20 0.58 -0.42 0.28 1999 1.22 -0.20 0.40 -0.56 0.23
Carex leporina/feta 1995 -0.40 -0.20 -0.80 -0.60 0.17 1997 -0.40 -0.20 -0.63 2.10 0.35 1999 -0.40 -0.20 -0.80 -0.60 0.17
Carex sp. 1 1995 4.98 12.87 9.77 1.34 0.63 1997 21.52 3.73 27.37 2.10 0.61 1999 1.23 19.2 13.02 2.03 0.56
Deschampsia cespitosa 1995 1.92 2.21 1.01 2.46 0.21 1997 0.99b 2.35a 2.16ab 3.12a 0.10 1999 0.01b 1.74a 1.07ab 0.65ab 0.08
Juncus tenuis 1995 43.74 20.12 16.25 5.73 0.35 1997 4.74a 1.93ab 6.08a 0.93b 0.02 1999 13.14 2.57 12.38 2.60 0.26
Group total 1995 6.29 2.62 2.08 2.71 0.58 1997 2.02 1.97 2.60 1.48 0.51 1999 1.73 1.41 1.47 0.76 0.18 Common native forbs Cover Eriophyllum lanatum 1995 0.00 0.12 0.00 -0.16 0.36 1997 0.00 0.47 0.00 -0.10 0.13 1999 0.47 0.55 0.00 0.35 0.89
Lotus purshiana 1995 -0.47 0.34 -0.43 -0.44 0.30 1997 -0.04a 14.83a -0.52ab -0.64b 0.09 1999 -0.15 3.88 -0.16 -0.41 0.77
Sidalcea cusickii ssp.purpurea 1995 0.00 0.00 -0.30 0.00 0.13
Controlling Woody Vegetation Page 34 Clark and Wilson
1997 0.00 0.00 -0.10 0.00 1.00 1999 0.00 0.00 0.00 0.00 1.00
Veronica scutellata 1995 -0.23 -0.24 -0.29 -0.39 0.89 1997 0.21ab 0.79a 0.98a -0.54b 0.08 1999 -0.13 0.00 -0.26 -0.36 0.81
Group total 1995 -0.41 -0.22 -0.57 -0.56 0.35 1997 0.08a 1.67a -0.14ab -0.77b 0.04 1999 -0.25 0.04 -0.25 -0.56 0.13 Number of inflorescences Sisyrinchium idahoensis 1995 0.00 0.00 0.00 0.00 1997 0.00 0.00 2.47 0.00 0.43 1999 6.27 0.00 4.27 0.00 0.11
Common non-native graminoids Number of inflorescences Anthoxanthum odoratum 1995 1.04 2.66 0.64 -0.27 0.45 1997 4.67ab 0.44bc 4.80a -0.25c 0.02 1999 1.39 0.80 13.39 3.22 0.33
Holcus lanatus 1995 -0.71 1.34 4.45 -0.38 0.45 1997 -0.72b 5.46a 11.50a -0.80b 0.01 1999 -0.90b 6.98a 1.21a -0.19a 0.02
Group total 1995 -0.54 -0.21 0.19 -0.52 0.68 1997 0.45cb 2.79a 1.92ab -0.68c 0.03 1999 -0.55b 0.59ab 1.47a -0.18b 0.01 Common non-native forbs Cover Centaurium umbellatum 1995 -0.40 -0.27 -0.37 -0.77 0.64 1997 0.22 1.97 2.01 3.41 0.42 1999 -0.40 -0.05 -0.46 -0.63 0.62
Chrysanthemum leucanthemum 1995 -0.35b 0.00a 0.00a 0.00a 0.10 1997 -0.28b 0.00a 0.00a 0.00a 0.10 1999 0.03 0.00 0.00 1.13 0.92
Hypericum perforatum 1995 -0.99 -0.89 -0.08 -0.60 0.25 1997 -0.89b -0.78b -0.24a 0.92a 0.00 1999 -1.00 -0.94 -0.47 -0.88 0.19
Lysimachia nummularia 1995 0.00 -0.14 0.00 -0.11 0.42 1997 0.00 -0.10 0.00 -0.06 0.42 1999 0.00 -0.13 0.00 -0.14 0.43
Vicia sativa 1995 0.00a 0.07a 0.13a -0.60b 0.03 1997 4.80 0.03 1.30 -0.20 0.45 1999 0.00 4.83 7.13 -0.05 0.30
Vicia tetrasperma 1995 -0.65ab -0.30a -0.15a -0.91b 0.07 1997 -0.03 1.52 0.39 0.35 0.68 1999 -0.04 3.15 3.91 1.19 0.37
Group total 1995 -0.89 -0.78 -0.48 -0.77 0.24 1997 -0.30b -0.45b 0.43a 0.28a 0.03
Controlling Woody Vegetation Page 35 Clark and Wilson
1999 -0.44 -0.11 2.32 -0.15 0.08 Overall cover (all species) 1995 0.01b 0.04b 0.32a -0.12b 0.01 1997 0.41 0.42 0.60 0.31 0.71 1999 0.33ab 0.42a 0.72a -0.09b 0.04
Controlling Woody Vegetation Page 36 Clark and Wilson
Table 5. Summary of the effectiveness of three management treatments (prescribed burning, hand-removal of woody species, and mowing with removal of cut material) in fulfilling management objectives for native wetland prairies in the Willamette Valley in 1999 based on results of three treatment periods from 1994 to 1998.
Restoration Objective Treatments
Burning Hand-removal of woody Mowing with removal species of cut material
Reduce abundance of YES YES NO woody species (decreased) (decreased) (no significant change)
Reduce or not promote abundance of non-native herbaceous species: - Graminoid inflorescences YES YES NO (no significant change) (no significant change) (increased)
- Forb cover YES YES YES (no significant change) (no significant change) (no significant change)
Promote or maintain abundance of native herbaceous species: - Graminoid cover YES YES YES (no significant change) (no significant change) (no significant change)
- Graminoid inflorescences NO YES YES (no significant change)* (no significant change) (no significant change)
- Forb cover YES YES YES (no significant change) (no significant change) (no significant change) * with important exception of reduction in the number of inflorescences for Deschampsia cespitosa
Controlling Woody Vegetation Page 37 Clark and Wilson